264 



SCIENCE. 



[N. S. Vol. XX. No. 504. 



ever else might be said of them, were at 

 least pieces of matter, and, like other pieces 

 of matter, possessed those 'primary' qual- 

 ities supposed to be characteristic of all 

 matter, whether found in large masses or 

 in small. 



But the electric theory which we have 

 been considering carries us into a new re- 

 gion altogether. It does not confine itself 

 to accounting for the secondary qualities 

 by the primary, or the behavior of matter 

 in bulk by the behavior of matter in atoms ; 

 it analyses matter, whether molar or molec- 

 ular, into something which is not matter at 

 all. The atom is now no more than the 

 relatively vast theater of operations in 

 which minute monads perform their or- 

 derly evolutions; while the monads them- 

 selves are not regarded as units of matter, 

 but as units of electricity; so that matter 

 is not merely explained, but is explained 

 away. 



Now the point to which I desire to call 

 attention is not to be sought in the great 

 divergence between matter as thus con- 

 ceived by the physicist and matter as the 

 ordinary man supposes himself to know it, 

 between matter as it is perceived and mat- 

 ter as it really is, but to the fact that the 

 first of these two quite inconsistent views 

 is wholly based on the second. 



This is surely something of a paradox. 

 We claim to found all our scientifie opin- 

 ions on experience ; and the experience on 

 which we foi^nd our theories of the physical 

 universe is our sense-perception of that 

 universe. That is experience; and in this 

 region of belief there is no other. Yet the 

 conclusions which thus profess to be en- 

 tirely founded upon experience are to all 

 appearance fundamentally opposed to it; 

 our knowledge of reality is based upon 

 illusion, and the very conceptions we use 

 in describing it to others, or in thinking of 

 it ourselves, are abstracted from anthro- 

 pomorphic fancies, which science forbids us 



to believe and nature compels us to employ. 

 We here touch the fringe of a series of 

 problems with which inductive logic ought 

 to deal, but which that most iinsatisfactory 

 branch of philosophy has systematically 

 ignored. This is no fault of men of sci- 

 ence. They are occupied in the task of 

 making discoveries, not in that of analyz- 

 ing the fundamental presuppositions which 

 the very possibility of making discoveries 

 implies. Neither is it the fault of trans- 

 cendental metaphysicians. Their specu- 

 lations flourish on a different level of 

 thought; their interest in a philosophy of 

 nature is lukewarm ; and howsoever the 

 questions in which they are chiefly con- 

 cerned be answered, it is by no means cer- 

 tain that the answers will leave the humbler 

 difficulties at which I have hinted either 

 nearer to or further from a solution. But 

 though men of science and idealists stand 

 acquitted, the same can hardly be said of 

 empirical philosophers. So far from solv- 

 ing the problem, they seem scarcely to have 

 understood that there was a problem to be 

 solved. Led astray by a misconception to 

 which I have already referred ; believing 

 that science was concerned only with (so- 

 called) 'phenomena,' that it had done all 

 that it could be asked to do if it accounted 

 for the sequence of our individual sensa- 

 tions, that it was concerned only with the 

 'laws of nature,' and not with the inner 

 character of physical reality; disbelieving, 

 indeed, that any such physical reality does 

 in truth exist;— it has never felt called 

 upon seriously to consider what are the 

 actual methods by which science attains its 

 I'esults, and how those methods are to be 

 justified. If anyone, for example, will 

 take up Mill's logic, Avith its 'seciuences 

 and co-existences between phenomena,' its 

 ' method of difference, ' its ' method of agree- 

 ment, ' and the rest ; if he will then compare 

 the actual doctrines of science with this 

 version of the mode in which those doc- 



